| { |
| "type": "lite", |
| "n_obs_steps": 1, |
| "input_features": { |
| "observation.images.wrist_image": { |
| "type": "VISUAL", |
| "shape": [ |
| 256, |
| 256, |
| 3 |
| ] |
| }, |
| "observation.images.image": { |
| "type": "VISUAL", |
| "shape": [ |
| 256, |
| 256, |
| 3 |
| ] |
| }, |
| "observation.state": { |
| "type": "STATE", |
| "shape": [ |
| 8 |
| ] |
| } |
| }, |
| "output_features": { |
| "action": { |
| "type": "ACTION", |
| "shape": [ |
| 7 |
| ] |
| } |
| }, |
| "device": "cuda", |
| "use_amp": false, |
| "push_to_hub": true, |
| "repo_id": "peeeeeter/libero_test", |
| "private": null, |
| "tags": null, |
| "license": null, |
| "pretrained_path": null, |
| "chunk_size": 100, |
| "n_action_steps": 100, |
| "normalization_mapping": { |
| "VISUAL": "IDENTITY", |
| "STATE": "MEAN_STD", |
| "ACTION": "MEAN_STD" |
| }, |
| "vision_backbone": "resnet18", |
| "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", |
| "replace_final_stride_with_dilation": false, |
| "pre_norm": false, |
| "dim_model": 512, |
| "n_heads": 8, |
| "dim_feedforward": 3200, |
| "feedforward_activation": "relu", |
| "n_encoder_layers": 4, |
| "decoder_type": "mlp", |
| "diffusion_steps": 10, |
| "n_decoder_layers": 1, |
| "use_vae": true, |
| "latent_dim": 32, |
| "n_vae_encoder_layers": 4, |
| "temporal_ensemble_coeff": null, |
| "dropout": 0.1, |
| "kl_weight": 10.0, |
| "diffusion_loss_weight": 1.0, |
| "diff_hidden_dim": 128, |
| "use_language": true, |
| "language_fusion": "late", |
| "language_model_type": "qwen", |
| "freeze_language_model": true, |
| "freeze_image_model": false, |
| "optimizer_lr": 1e-05, |
| "optimizer_weight_decay": 0.0001, |
| "optimizer_lr_backbone": 1e-05 |
| } |
